In vitro reconstitution of lateral to end-on conversion of kinetochore-microtubule attachments

Abstract

During mitosis, kinetochores often bind to the walls of spindle microtubules, but these lateral interactions are then converted into a different binding mode in which microtubule plus-ends are embedded at kinetochores, forming dynamic "end-on" attachments. This remarkable configuration allows continuous addition or loss of tubulin subunits from the kinetochore-bound microtubule ends, concomitant with movement of the chromosomes. Here, we describe novel experimental assays for investigating this phenomenon using a well-defined in vitro reconstitution system visualized by fluorescence microscopy. Our assays take advantage of the kinetochore kinesin CENP-E, which assists in microtubule end conversion in vertebrate cells. In the experimental setup, CENP-E is conjugated to coverslip-immobilized microbeads coated with selected kinetochore components, creating conditions suitable for microtubule gliding and formation of either static or dynamic end-on microtubule attachment. This system makes it possible to analyze, in a systematic and rigorous manner, the molecular friction generated by the microtubule wall-binding proteins during lateral transport, as well as the ability of these proteins to establish and maintain association with microtubule plus-end, providing unique insights into the specific activities of various kinetochore components.

Keywords: Fluorescence microscopy; Gliding assay; Kinesin CENP-E; Microtubule dynamics; Microtubule end-coupling; Microtubule-associated proteins.

FIG. 1

Gliding assay with CENP-E motor. (A) Velocity of MT gliding (mean±SEM) as a function of coverslip brightness, measured via GFP fluorescence. Coverslip was coated with truncated CENP-E-GFP according to steps 1–7 in Section 3.1. CENP-E-GFP density was varied by using different concentrations of anti-GFP antibody and excess soluble CENP-E-GFP motor. Open and filled circles correspond to the translational velocities of linearly gliding MTs and pivoting MTs, respectively. (B) MT gliding velocity vs MT length (mean±SD) in experiments using our “modified” assay described in Section 3.1 or a “regular” gliding protocol in which CENP-E motors and antibodies were adsorbed on the coverslip lacking the BSA-neutravidin coating. (C) Representative images of two MTs moving over coverslips with either high (>10³a.u.) or low (<10³a.u.) motor densities. Numbers indicate imaging time (in seconds). Bottom images show maximum intensity projections, illustrating predominantly linear motion (left) or translocation with pivoting around a fixed point (arrow), indicative of a single motor interaction.

FIG. 2

Interactions between a kinetochore MAP and the walls of the CENP-E-transported microtubules. (A) Schematic representation of the MT gliding assay using truncated CENP-E-Myc and a kinetochore MAP tagged with GFP. Proteins were immobilized on the coverslip with anti-Myc and anti-GFP antibodies, shown as black and green Ys, respectively. (B) Schematic of perfusion through the flow chamber used in this assay. (C) Maximum projection of a 100-frame time series of gliding MTs acquired at 1 frame/s. Colors represent different times, as indicated by the color-coded scale below. (D) Representative kymographs of three MTs gliding on surfaces coated with CENP-E and a kinetochore MAP (Ndc80 protein complex) at different surface densities.

FIG. 3

Transition from lateral to end microtubule binding and coupling at the dynamic microtubule plus-end. (A) Schematic representation of the end-conversion assay (not to scale), in which stabilized MTs glide on the surface of coverslip-immobilized beads. (B) Schematics of perfusion through the reusable flow chamber used in this assay. (C) Selected images of taxol-stabilized MTs (red) gliding over beads (green) coated with CENP-E kinesin and kinetochore MAP. The number at the top left corner of each frame indicates the acquisition time (in minutes) of that frame. Landing of a second MT is marked by white triangle. (D) Representative kymograph of a MT obtained with beads coated with CENP-E kinesin and a kinetochore MAP. The MT glides then stays attached to the bead at the MT plus-end. (E) Kymograph of the fluorescent MT seed repeatedly moving away from and back to the bead as MT polymerizes (P) and depolymerizes (D) in the presence of soluble unlabeled tubulin.